biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[AOT compilation] support torch.compile inductor artifacts in VllmCompiledFunction (#25205)

Browse Source 
Signed-off-by: dolpm <34420038+dolpm@users.noreply.github.com>
This commit is contained in:
dolpm
2026-01-20 11:45:59 -08:00
committed by GitHub
parent 193069d129
commit 7c5dedc247
8 changed files with 1169 additions and 113 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tools/pre_commit/check_pickle_imports.py
View File

@@ -21,6 +21,7 @@ ALLOWED_FILES = {
"vllm/transformers_utils/config.py",
"vllm/model_executor/models/registry.py",
"vllm/compilation/caching.py",
"vllm/compilation/piecewise_backend.py",
"vllm/distributed/utils.py",
"vllm/distributed/parallel_state.py",
"vllm/distributed/device_communicators/all_reduce_utils.py",
@@ -30,6 +31,7 @@ ALLOWED_FILES = {
"tests/multimodal/media/test_base.py",
"tests/tokenizers_/test_hf.py",
"tests/utils_/test_hashing.py",
"tests/compile/test_aot_compile.py",
"benchmarks/kernels/graph_machete_bench.py",
"benchmarks/kernels/benchmark_lora.py",
"benchmarks/kernels/benchmark_machete.py",